We have studied the mechanisms regulating m2- and m3-muscarinic and serotonergic 5-HT2 receptors in cultured cerebellar granule cells. The selectivity of antagonists to bind to the two muscarinic receptor subtypes is markedly reduced by cell disruption to prepare membranes for binding assays. The m2-receptor-selective antagonists AF-DX 116 and methoctramine, and two m3-selective antagonists 4-DAMP and pFHHSiD, have decreased affinities in membrane preparations for the receptor subtype for which they had higher affinity to start with. This caused a net decrease in the selectivity of the compounds. The reason for the decrease in selectivity was not due to disruption of interactions between the receptors with the cytoskeleton or exposure to extracellular levels of calcium. The antagonist-specific changes in subtype selectivity may indicate that these compounds affect the receptor-G-protein coupling in intact cells. For the agonist-induced up-regulation of 5-HT2A receptors we found that this process required de novo and protein synthesis, but was independent of the receptor-mediated production of inositol trisphosphate and diacylglycerol and activation of protein kinase C. The increase in 5-HT2A receptor protein has been verified by immunoblotting. The agonist up-regulation required receptor-mediated Ca2+ influx and a calmodulin-dependent pathway, possibly involving Ca2+/calmodium kinase. Additionally, the induction of immediate early genes and an increase in AP-1 and CREB binding were associated with 5-HT2A up-regulation, impli- cating their roles for transcriptional regulation. The agonist receptor up-regulation involves no change in Gq/11 alpha-subunit proteins. A number of 5-HT2A receptor antagonists were also found to up-regulate 5- HT2A receptor sites, suggesting that these drugs have intrinsic activity. This intrinsic activity may be related to the therapeutic actions of some 5-HT2A receptor antagonists.